Biphasic modulation of NOS expression, protein and nitrite products by hydroxocobalamin underlies its protective effect in endotoxemic shock: downstream regulation of COX-2, IL-1β, TNF-α, IL-6, and HMGB1 expression

Abstract

Background: NOS/•NO inhibitors are potential therapeutics for sepsis, yet they increase clinical mortality. However, there has been no in vivo investigation of the (in vitro) •NO scavenger, cobalamin's (Cbl) endogenous effects on NOS/•NO/inflammatory mediators during the immune response to sepsis.

Methods: We used quantitative polymerase chain reaction (qPCR), ELISA, Western blot, and NOS Griess assays, in a C57BL/6 mouse, acute endotoxaemia model.

Results: During the immune response, pro-inflammatory phase, parenteral hydroxocobalamin (HOCbl) treatment partially inhibits hepatic, but not lung, iNOS mRNA and promotes lung eNOS mRNA, but attenuates the LPS hepatic rise in eNOS mRNA, whilst paradoxically promoting high iNOS/eNOS protein translation, but relatively moderate •NO production. HOCbl/NOS/•NO regulation is reciprocally associated with lower 4 h expression of TNF-α, IL-1β, COX-2, and lower circulating TNF-α, but not IL-6. In resolution, 24 h after LPS, HOCbl completely abrogates a major late mediator of sepsis mortality, high mobility group box 1 (HMGB1) mRNA, inhibits iNOS mRNA, and attenuates LPS-induced hepatic inhibition of eNOS mRNA, whilst showing increased, but still moderate, NOS activity, relative to LPS only. experiments (LPS+D-Galactosamine) HOCbl afforded significant, dose-dependent protection in mice.

Conclusions: HOCbl produces a complex, time- and organ-dependent, selective regulation of NOS/•NO during endotoxaemia, corollary regulation of downstream inflammatory mediators, and increased survival. This merits clinical evaluation.

Figure 1

The structure of cobalamin. X = the principal ligands for the cobalt atom, in the upper, β axial position.

Figure 2

HOCbl protects mice against experimental endotoxaemia. In (a) and (b) mice (n = 8 and 7, resp.) were treated with HOCbl or GSCbl or NAC-Cbl following two distinct protocols, in addition to being injected with LPS (0.1 mg/kg i.p.) + D-Galactosamine (1 g/kg i.p.). Impact on lethality was followed for a total of 5 days. (a) Cobalamins were administered at a dose of 0.2 mg/kg i.p. −1 h, +1 h, +2 h, +6 h, and +22 h and compared to controls given LPS+D-Galactosamine injection alone (Time 0 h). *P < 0.05 for GSCbl/NAC-Cbl and + for HOCbl when tested against LPS+D = Gal (n = 9), using the Chi-square or Kaplan-Meier tests. (b) Cobalamins were administered at a dose of 40 mg/kg i.p. +2 h and +6 h and compared to controls given LPS + D-Galactosamine injection alone (Time 0 h). *P < 0.01 for NAC-Cbl and + for GSCbl/HOCbl, when tested against LPS+D-Gal (n = 9), using the Chi-square or Kaplan-Meier tests.

Figure 3

Ultra-high dose HOCbl consistently improves survival in experimental endotoxaemia. Mice (n = 12 per group: 2 Cbl active treatment groups plus 1 LPS-only group) were treated with HOCbl, following two distinct protocols, +40 mg/kg at 2 h and +4 h post LPS+D-Gal, and +80 mg/kg at only +2 h following LPS (0.1 mg/kg i.p.) + D-Galactosamine (1 g/kg i.p.). Impact on lethality was followed for a total of 5 days. *P < 0.05 for HOCbl at both doses, when tested against LPS+D-Gal (n = 12), using the Chi-square test. *P < 0.01 for HOCbl (80 mg/kg) by using the Kaplan-Meier test.

Figure 4

HOCbl selectively modulates NOS enzymes and inhibits COX-2, IL-1β mRNA, in lung and liver, at 4 h following LPS-induced endotoxaemia. Mice (n = 5) were treated with HOCbl, at the low dose protocol (0.2 mg/kg i.p.) and compared to LPS only (0.1 mg/kg i.p. at time 0 h). Organs (lung and liver) were harvested at 4 h after LPS and gene expression was quantified in tissue extracts by real-time PCR, using GAPDH and RPL32 as internal standards. ((a), (b)) eNOS mRNA data; ((c), (d)) iNOS mRNA data; ((e), (f)) COX-2 mRNA data; ((g), (h)) IL-1β mRNA data. Values are a mean ± SEM of triplicate observations. *P < 0.05 versus vehicle (PBS-treated) control; ⁺ P < 0.05 versus LPS-only group.

Figure 5

Early (4 h) promotional effect of HOCbl on iNOS and eNOS protein expression in liver samples from endotoxemic mice: mice (n = 5) were treated with HOCbl at the low dose protocol (0.2 mg/kg i.p.) at –1 h, +1 h, and +2 h and challenged with LPS (0.1 mg/kg i.p.) at time 0. Liver samples were collected and processed as described in Materials and Methods, for the Western blot analysis of iNOS and eNOS protein expression. Upper panels ((a), (b)) show membranes probed for iNOS, eNOS, with tubulin as the loading control; lower panels show densitometric analysis of the blots. Values are a mean ± SEM of triplicate experiments. *P < 0.05 versus vehicle (PBS-treated) control; ⁺ P < 0.05 versus LPS group.

Figure 6

Modulatory effects of HOCbl on NOS activity in tissue homogenates of endotoxemic mice at 4 h and 24 h. Liver ((a), (c)) and lung (b) tissue samples were collected 4 h and 24 h after LPS challenge (0.1 mg/kg; i.p.); two groups of mice (n = 5 per group) having previously been treated with HOCbl (resp., 0.2 mg/kg i.p. at −1 h, +1 h, and +2 h and at −1 h, +1 h, +2 h, +6 h, and +22 h) and also challenged with LPS at time 0. NOS activity at 4 h and 24 h was assayed as described in Materials and Methods, and shown as µmol of NO generated per µg protein. Values are a mean ± SEM of triplicate observations. *P < 0.05 versus vehicle (PBS-treated) control; ⁺ P < 0.05 versus LPS group.

Figure 7

The effects of thiol Cbls on activity of NOS. Liver tissue samples were collected 4 h after LPS challenge (0.1 mg/kg; i.p.); 4 groups of mice (n = 5 per group) having previously been treated with either LPS-only or HOCbl/or GSCbl or NAC-Cbl plus LPS at time 0 h (0.2 mg/kg i.p. at −1 h, +1 h, and +2 h and at −1 h, +1 h, +2 h, +6 h, and +22 h). NOS activity at 4 h was assayed as described in Materials and Methods, and shown as µmol of NO generated per µg protein. Values are a mean ± SEM of triplicate observations. *P < 0.05 versus vehicle (PBS-treated) control; ⁺ P < 0.05 versus LPS group.

Figure 8

Selective effects of HOCbl on IL-6 and TNF-alpha at 4 h following LPS-induced endotoxaemia. Mice (n = 5) were treated with HOCbl, at the low dose protocol (0.2 mg/kg i.p.) at −1 h, +1 h, and +2 h and compared to LPS (0.1 mg/kg i.p. at time 0). Organs (lung and liver) were harvested at 4 h subsequent to LPS, and gene expression quantified in tissue extracts by real-time PCR, using GAPDH and RPL32 as internal standards. ELISA was used to measure plasma levels of IL-6 (a) and TNF-α (b). Tissue expression of TNF-α mRNA in lung (c) and liver (d) extracts. Data are a mean ± SEM of 5 mice per group. *P < 0.05 versus vehicle (PBS-treated) control; ⁺ P < 0.05 versus LPS group.

Figure 9

HOCbl modulates eNOS/iNOS, COX-2, HMGB1 gene expression in lung and liver at 24 h subsequent to LPS-induced endotoxaemia. Mice (n = 5) were treated with HOCbl (0.2 mg/kg i.p low dose regimen (Table 1)) and compared to LPS 0.1 mg/kg (given i.p. at time 0 h). Organs (lung and liver) were harvested at 24 h after LPS and gene expression was quantified in tissue extracts by real-time PCR, using GAPDH and RPL32 as internal standards, and PBS-injected group set as 1. ((a), (b)) eNOS mRNA data; ((c), (d)) iNOS mRNA data; ((e), (f)) COX-2 mRNA data; ((g), (h)) HMGB1 mRNA data. Values are a mean ± SEM of triplicate observations. *P < 0.05 versus vehicle (PBS-treated) control; ⁺ P < 0.05 versus LPS group.

Figure 10

Hypothetical scheme showing how cobalamins may modulate the NOS/^•NO species, both indirectly and directly, with downstream effects, during the course of the immune response. Immune challenge upregulates—via the AP-1/Ras signalling pathway—the circulating Cbl carrier proteins, haptocorrins, /HC-TCI & III (which, when desialylated, are taken up by the hepatic Ashwell receptor, thus transporting more MeCbl to the liver for increased synthesis of acute phase proteins). At the same time, the Cbl tissue carrier proteins, the transcobalamins, /TCII are also upregulated, as a result of crosstalk between the transcription factors, NF-κB and Sp-1, the TCII promoter. The increased supply of intracellular Cbl is partly converted to the active cofactors, AdoCbl and MeCbl, in a high AdoCbl to low MeCbl ratio. When methionine synthase/MS, along with SAM synthesis, is initially decreased by LPS challenge in activated monocytes/macrophages, methylmalonyl CoA mutase/MU activity takes precedence [91], triggering increased expression of TCII cell receptors. Part of the consequent extra AdoCbl synthesised may then be directed towards NOS catalysis. A combined proportional increase in activity of MU and, later, MS ensures that all the necessary components for NOS synthesis/assembly, substrates and cofactors, are in plentiful supply. This prevents excess generation of ROS and RNIS during NOS catalysis. These are further decreased because part of the ^•NO produced may combine with GSH, or other thiols, to form the more beneficial, antioxidant S-nitrosothiols/S-NOS. ^•NO is then released, as needed, in targeted amounts, by the action of GSH-dependent GSNO reductase and/or possibly by interaction with HOCbl, yielding GSCbl and ^•NO. Increasing levels of GSNO/^•NO rapidly downregulate IL-1β, COX-2, TNF-α, HMGB1 expression, and, at the conclusion of the second, anti-inflammatory phase of the immune response also inhibit iNOS, NF-κB, and down-regulate MU, MS, and CBS. The lower right part of the scheme shows a possible direct catalytic regulation of iNOS by AdoCbl's lower axial ligand base, the dimethylbenzimidazole (DMBI), and the adenosyl radical, generated after NOS enzyme-induced homolysis of the Co-C bond, which would reduce formation of RNIS/ROS, increase the ratio of ^•NO/GSNO to ONOO- and related species, and make NOS catalysis more productive, thereby lessening the duration of iNOS activity and ensuring the beneficial effects of ^•NO predominate (see [43]).